Osteoporosis is responsible for approximately 1.5 million fractures in the US per year, with 300,000 of these fractures occurring at the hip at a cost exceeding $17 billion. The NIH Consensus conference and the World Health Organization defined osteoporosis as ...compromised bone strength predisposing to an increased risk of fracture. A variety of genetic and environmental factors, as well as bone properties (geometric and material), contribute to the bone loss that is associated with osteoporosis, but the question remains as to which factors primarily contribute to fracture risk. While reduced bone mineral density (BMD) relative to young individuals is routinely used clinically to predict fracture risk, BMD is not a strong predictor, with the majority of fractures occurring in patients with BMD's above the osteoporotic threshold. We have recently shown by multiple logistic regression analysis that specific mineral and matrix properties assessed by Fourier transform infrared spectroscopic imaging (FTIRI) are predictive of fracture in postmenopausal women, while BMD is not significantly associated with fracture incidence. In a limited number of samples we have also shown that these FTIR parameters are correlated with nanomechanical properties. We hypothesize that variation in crystallinity (XST) and collagen maturity (XLR) partially explains the difference in incidence of fractures in individuals with similar BMDs. We further hypothesize that heterogeneity in these parameters is an additional determinant of fracture incidence, especially in trabecular bone. In the proposed studies we will test 4 hypotheses. 1) For any subject, FTIRI data obtained in the cortical and cancellous bone of the iliac crest (generally a non-fracturing site) is representative of that from sites that fracture (subtrochanter/greater trochanter). Further, the data are independent of the size of the biopsy as long as cortical and cancellous bone areas are included. This will be tested using multiple biopsies from cadavers and from clinic patients with fractures. Measures will include micro-CT analysis of BMD and architecture, and FTIRI. 2) Decreased heterogeneity in FTIRI mineral and matrix properties, in addition to increased XST and XLR, are predictive of fracture in humans. This will be tested by extending our logistic regression to heterogeneity parameters. Variation in tissue properties with age will be studied in patients with idiopathic juvenile osteoporosis. Tissue mechanical properties will be correlated with FTIRI data. 3) An anabolic agent (PTH) can restore mechanical properties, and the XST, XLR, and mineral and matrix heterogeneity in animal models as well as in osteoporotic humans. This will be tested by analyses of pre- and post- treatment human biopsies and by analyses in a sheep model. 4) XLR, which is altered in osteoporosis, is related to collagen orientation. As part of our continued parameter validation, this will be tested by comparing FTIRI and second harmonic heneration microscopy data. Testing of these four hypotheses will provide new insights into the efficacy of therapies and contribute to the understanding of factors leading to fracture.